models: A List of Available Models in train

In caret: Classification and Regression Training

Description

These models are included in the package via wrappers for train. Custom models can also be created. See the URL below.

Bagged CART (method = 'treebag')

For classification and regression using packages ipred and plyr with no tuning parameters

Bagged Flexible Discriminant Analysis (method = 'bagFDA')

For classification using packages earth and mda with tuning parameters:

• Product Degree (degree)

• Number of Terms (nprune)

Bagged Logic Regression (method = 'logicBag')

For classification and regression using package logicFS with tuning parameters:

• Maximum Number of Leaves (nleaves)

• Number of Trees (ntrees)

Bagged MARS (method = 'bagEarth')

For classification and regression using package earth with tuning parameters:

• Number of Terms (nprune)

• Product Degree (degree)

Bagged Model (method = 'bag')

For classification and regression using package caret with tuning parameters:

• Number of Randomly Selected Predictors (vars)

Bayesian Generalized Linear Model (method = 'bayesglm')

For classification and regression using package arm with no tuning parameters

Bayesian Regularized Neural Networks (method = 'brnn')

For regression using package brnn with tuning parameters:

• Number of Neurons (neurons)

Boosted Classification Trees (method = 'ada')

For classification using package ada with tuning parameters:

• Number of Trees (iter)

• Max Tree Depth (maxdepth)

• Learning Rate (nu)

Boosted Generalized Additive Model (method = 'gamboost')

For classification and regression using package mboost with tuning parameters:

• Number of Boosting Iterations (mstop)

• AIC Prune? (prune)

Boosted Generalized Linear Model (method = 'glmboost')

For classification and regression using package mboost with tuning parameters:

• Number of Boosting Iterations (mstop)

• AIC Prune? (prune)

Boosted Linear Model (method = 'bstLs')

For classification and regression using packages bst and plyr with tuning parameters:

• Number of Boosting Iterations (mstop)

• Shrinkage (nu)

Boosted Logistic Regression (method = 'LogitBoost')

For classification using package caTools with tuning parameters:

• Number of Boosting Iterations (nIter)

Boosted Smoothing Spline (method = 'bstSm')

For classification and regression using packages bst and plyr with tuning parameters:

• Number of Boosting Iterations (mstop)

• Shrinkage (nu)

Boosted Tree (method = 'blackboost')

For classification and regression using packages party, mboost and plyr with tuning parameters:

• Number of Trees (mstop)

• Max Tree Depth (maxdepth)

Boosted Tree (method = 'bstTree')

For classification and regression using packages bst and plyr with tuning parameters:

• Number of Boosting Iterations (mstop)

• Max Tree Depth (maxdepth)

• Shrinkage (nu)

C4.5-like Trees (method = 'J48')

For classification using package RWeka with tuning parameters:

• Confidence Threshold (C)

C5.0 (method = 'C5.0')

For classification using packages C50 and plyr with tuning parameters:

• Number of Boosting Iterations (trials)

• Model Type (model)

• Winnow (winnow)

CART (method = 'rpart')

For classification and regression using package rpart with tuning parameters:

• Complexity Parameter (cp)

CART (method = 'rpart2')

For classification and regression using package rpart with tuning parameters:

• Max Tree Depth (maxdepth)

Conditional Inference Random Forest (method = 'cforest')

For classification and regression using package party with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

Conditional Inference Tree (method = 'ctree')

For classification and regression using package party with tuning parameters:

• 1 - P-Value Threshold (mincriterion)

Conditional Inference Tree (method = 'ctree2')

For classification and regression using package party with tuning parameters:

• Max Tree Depth (maxdepth)

Cost-Sensitive C5.0 (method = 'C5.0Cost')

For classification using packages C50 and plyr with tuning parameters:

• Number of Boosting Iterations (trials)

• Model Type (model)

• Winnow (winnow)

• Cost (cost)

Cost-Sensitive CART (method = 'rpartCost')

For classification using package rpart with tuning parameters:

• Complexity Parameter (cp)

• Cost (Cost)

Cubist (method = 'cubist')

For regression using package Cubist with tuning parameters:

• Number of Committees (committees)

• Number of Instances (neighbors)

Elasticnet (method = 'enet')

For regression using package elasticnet with tuning parameters:

• Fraction of Full Solution (fraction)

• Weight Decay (lambda)

Extreme Learning Machine (method = 'elm')

For classification and regression using package elmNN with tuning parameters:

• Number of Hidden Units (nhid)

• Activation Function (actfun)

Factor-Based Linear Discriminant Analysis (method = 'RFlda')

For classification using package HiDimDA with tuning parameters:

• Number of Factors (q)

Flexible Discriminant Analysis (method = 'fda')

For classification using packages earth and mda with tuning parameters:

• Product Degree (degree)

• Number of Terms (nprune)

Gaussian Process (method = 'gaussprLinear')

For classification and regression using package kernlab with no tuning parameters

Gaussian Process with Polynomial Kernel (method = 'gaussprPoly')

For classification and regression using package kernlab with tuning parameters:

• Polynomial Degree (degree)

• Scale (scale)

Gaussian Process with Radial Basis Function Kernel (method = 'gaussprRadial')

For classification and regression using package kernlab with tuning parameters:

• Sigma (sigma)

Generalized Additive Model using LOESS (method = 'gamLoess')

For classification and regression using package gam with tuning parameters:

• Span (span)

• Degree (degree)

Generalized Additive Model using Splines (method = 'gam')

For classification and regression using package mgcv with tuning parameters:

• Feature Selection (select)

• Method (method)

Generalized Additive Model using Splines (method = 'gamSpline')

For classification and regression using package gam with tuning parameters:

• Degrees of Freedom (df)

Generalized Linear Model (method = 'glm')

For classification and regression with no tuning parameters

Generalized Linear Model with Stepwise Feature Selection (method = 'glmStepAIC')

For classification and regression using package MASS with no tuning parameters

Generalized Partial Least Squares (method = 'gpls')

For classification using package gpls with tuning parameters:

• Number of Components (K.prov)

glmnet (method = 'glmnet')

For classification and regression using package glmnet with tuning parameters:

• Mixing Percentage (alpha)

• Regularization Parameter (lambda)

Greedy Prototype Selection (method = 'protoclass')

For classification using packages proxy and protoclass with tuning parameters:

• Ball Size (eps)

• Distance Order (Minkowski)

Heteroscedastic Discriminant Analysis (method = 'hda')

For classification using package hda with tuning parameters:

• Gamma (gamma)

• Lambda (lambda)

• Dimension of the Discriminative Subspace (newdim)

High Dimensional Discriminant Analysis (method = 'hdda')

For classification using package HDclassif with tuning parameters:

• Threshold (threshold)

• Model Type (model)

Independent Component Regression (method = 'icr')

For regression using package fastICA with tuning parameters:

• Number of Components (n.comp)

k-Nearest Neighbors (method = 'kknn')

For classification and regression using package kknn with tuning parameters:

• Max. Number of Neighbors (kmax)

• Distance (distance)

• Kernel (kernel)

k-Nearest Neighbors (method = 'knn')

For classification and regression with tuning parameters:

• Number of Neighbors (k)

Learning Vector Quantization (method = 'lvq')

For classification using package class with tuning parameters:

• Codebook Size (size)

• Number of Prototypes (k)

Least Angle Regression (method = 'lars')

For regression using package lars with tuning parameters:

• Fraction (fraction)

Least Angle Regression (method = 'lars2')

For regression using package lars with tuning parameters:

• Number of Steps (step)

Least Squares Support Vector Machine (method = 'lssvmLinear')

For classification using package kernlab with no tuning parameters

Least Squares Support Vector Machine with Polynomial Kernel (method = 'lssvmPoly')

For classification using package kernlab with tuning parameters:

• Polynomial Degree (degree)

• Scale (scale)

Least Squares Support Vector Machine with Radial Basis Function Kernel (method = 'lssvmRadial')

For classification using package kernlab with tuning parameters:

• Sigma (sigma)

Linear Discriminant Analysis (method = 'lda')

For classification using package MASS with no tuning parameters

Linear Discriminant Analysis (method = 'lda2')

For classification using package MASS with tuning parameters:

• Number of Discriminant Functions (dimen)

Linear Discriminant Analysis with Stepwise Feature Selection (method = 'stepLDA')

For classification using packages klaR and MASS with tuning parameters:

• Maximum Number of Variables (maxvar)

• Search Direction (direction)

Linear Regression (method = 'lm')

For regression with no tuning parameters

Linear Regression with Backwards Selection (method = 'leapBackward')

For regression using package leaps with tuning parameters:

• Maximum Number of Predictors (nvmax)

Linear Regression with Forward Selection (method = 'leapForward')

For regression using package leaps with tuning parameters:

• Maximum Number of Predictors (nvmax)

Linear Regression with Stepwise Selection (method = 'leapSeq')

For regression using package leaps with tuning parameters:

• Maximum Number of Predictors (nvmax)

Linear Regression with Stepwise Selection (method = 'lmStepAIC')

For regression using package MASS with no tuning parameters

Logic Regression (method = 'logreg')

For classification and regression using package LogicReg with tuning parameters:

• Maximum Number of Leaves (treesize)

• Number of Trees (ntrees)

Logistic Model Trees (method = 'LMT')

For classification using package RWeka with tuning parameters:

• Number of Iteratons (iter)

Maximum Uncertainty Linear Discriminant Analysis (method = 'Mlda')

For classification using package HiDimDA with no tuning parameters

Mixture Discriminant Analysis (method = 'mda')

For classification using package mda with tuning parameters:

• Number of Subclasses Per Class (subclasses)

Model Averaged Neural Network (method = 'avNNet')

For classification and regression using package nnet with tuning parameters:

• Number of Hidden Units (size)

• Weight Decay (decay)

• Bagging (bag)

Model Rules (method = 'M5Rules')

For regression using package RWeka with tuning parameters:

• Pruned (pruned)

• Smoothed (smoothed)

Model Tree (method = 'M5')

For regression using package RWeka with tuning parameters:

• Pruned (pruned)

• Smoothed (smoothed)

• Rules (rules)

Multi-Layer Perceptron (method = 'mlp')

For classification and regression using package RSNNS with tuning parameters:

• Number of Hidden Units (size)

Multi-Layer Perceptron (method = 'mlpWeightDecay')

For classification and regression using package RSNNS with tuning parameters:

• Number of Hidden Units (size)

• Weight Decay (decay)

Multivariate Adaptive Regression Spline (method = 'earth')

For classification and regression using package earth with tuning parameters:

• Number of Terms (nprune)

• Product Degree (degree)

Multivariate Adaptive Regression Splines (method = 'gcvEarth')

For classification and regression using package earth with tuning parameters:

• Product Degree (degree)

Naive Bayes (method = 'nb')

For classification using package klaR with tuning parameters:

• Laplace Correction (fL)

• Distribution Type (usekernel)

Nearest Shrunken Centroids (method = 'pam')

For classification using package pamr with tuning parameters:

• Shrinkage Threshold (threshold)

Neural Network (method = 'neuralnet')

For regression using package neuralnet with tuning parameters:

• Number of Hidden Units in Layer 1 (layer1)

• Number of Hidden Units in Layer 2 (layer2)

• Number of Hidden Units in Layer 3 (layer3)

Neural Network (method = 'nnet')

For classification and regression using package nnet with tuning parameters:

• Number of Hidden Units (size)

• Weight Decay (decay)

Neural Networks with Feature Extraction (method = 'pcaNNet')

For classification and regression using package nnet with tuning parameters:

• Number of Hidden Units (size)

• Weight Decay (decay)

Oblique Random Forest (method = 'ORFlog')

For classification using package obliqueRF with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

Oblique Random Forest (method = 'ORFpls')

For classification using package obliqueRF with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

Oblique Random Forest (method = 'ORFridge')

For classification using package obliqueRF with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

Oblique Random Forest (method = 'ORFsvm')

For classification using package obliqueRF with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

Oblique Trees (method = 'oblique.tree')

For classification using package oblique.tree with tuning parameters:

• Oblique Splits (oblique.splits)

• Variable Selection Method (variable.selection)

Parallel Random Forest (method = 'parRF')

For classification and regression using package randomForest with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

partDSA (method = 'partDSA')

For classification and regression using package partDSA with tuning parameters:

• Number of Terminal Partitions (cut.off.growth)

• Minimum Percent Difference (MPD)

Partial Least Squares (method = 'kernelpls')

For classification and regression using package pls with tuning parameters:

• Number of Components (ncomp)

Partial Least Squares (method = 'pls')

For classification and regression using package pls with tuning parameters:

• Number of Components (ncomp)

Partial Least Squares (method = 'simpls')

For classification and regression using package pls with tuning parameters:

• Number of Components (ncomp)

Partial Least Squares (method = 'widekernelpls')

For classification and regression using package pls with tuning parameters:

• Number of Components (ncomp)

Penalized Discriminant Analysis (method = 'pda')

For classification using package mda with tuning parameters:

• Shrinkage Penalty Coefficient (lambda)

Penalized Discriminant Analysis (method = 'pda2')

For classification using package mda with tuning parameters:

• Degrees of Freedom (df)

Penalized Linear Discriminant Analysis (method = 'PenalizedLDA')

For classification using packages penalizedLDA and plyr with tuning parameters:

• L1 Penalty (lambda)

• Number of Discriminant Functions (K)

Penalized Linear Regression (method = 'penalized')

For regression using package penalized with tuning parameters:

• L1 Penalty (lambda1)

• L2 Penalty (lambda2)

Penalized Logistic Regression (method = 'plr')

For classification using package stepPlr with tuning parameters:

• L2 Penalty (lambda)

• Complexity Parameter (cp)

Penalized Multinomial Regression (method = 'multinom')

For classification using package nnet with tuning parameters:

• Weight Decay (decay)

Polynomial Kernel Regularized Least Squares (method = 'krlsPoly')

For regression using package KRLS with tuning parameters:

• Regularization Parameter (lambda)

• Polynomial Degree (degree)

Principal Component Analysis (method = 'pcr')

For regression using package pls with tuning parameters:

• Number of Components (ncomp)

Projection Pursuit Regression (method = 'ppr')

For regression with tuning parameters:

• Number of Terms (nterms)

Quadratic Discriminant Analysis (method = 'qda')

For classification using package MASS with no tuning parameters

Quadratic Discriminant Analysis with Stepwise Feature Selection (method = 'stepQDA')

For classification using packages klaR and MASS with tuning parameters:

• Maximum Number of Variables (maxvar)

• Search Direction (direction)

Quantile Random Forest (method = 'qrf')

For regression using package quantregForest with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

Quantile Regression Neural Network (method = 'qrnn')

For regression using package qrnn with tuning parameters:

• Number of Hidden Units (n.hidden)

• Weight Decay (penalty)

• Bagged Models? (bag)

Radial Basis Function Kernel Regularized Least Squares (method = 'krlsRadial')

For regression using packages KRLS and kernlab with tuning parameters:

• Regularization Parameter (lambda)

• Sigma (sigma)

Radial Basis Function Network (method = 'rbf')

For classification using package RSNNS with tuning parameters:

• Number of Hidden Units (size)

Radial Basis Function Network (method = 'rbfDDA')

For classification and regression using package RSNNS with tuning parameters:

• Activation Limit for Conflicting Classes (negativeThreshold)

Random Ferns (method = 'rFerns')

For classification using package rFerns with tuning parameters:

• Fern Depth (depth)

Random Forest (method = 'rf')

For classification and regression using package randomForest with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

Random Forest by Randomization (method = 'extraTrees')

For classification and regression using package extraTrees with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

• Number of Random Cuts (numRandomCuts)

Random Forest with Additional Feature Selection (method = 'Boruta')

For classification and regression using packages Boruta and randomForest with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

Random k-Nearest Neighbors (method = 'rknn')

For classification and regression using package rknn with tuning parameters:

• Number of Neighbors (k)

• Number of Randomly Selected Predictors (mtry)

Random k-Nearest Neighbors with Feature Selection (method = 'rknnBel')

For classification and regression using packages rknn and plyr with tuning parameters:

• Number of Neighbors (k)

• Number of Randomly Selected Predictors (mtry)

• Number of Features Dropped (d)

Regularized Discriminant Analysis (method = 'rda')

For classification using package klaR with tuning parameters:

• Gamma (gamma)

• Lambda (lambda)

Regularized Random Forest (method = 'RRF')

For classification and regression using packages randomForest and RRF with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

• Regularization Value (coefReg)

• Importance Coefficient (coefImp)

Regularized Random Forest (method = 'RRFglobal')

For classification and regression using package RRF with tuning parameters:

• Number of Randomly Selected Predictors (mtry)

• Regularization Value (coefReg)

Relaxed Lasso (method = 'relaxo')

For regression using packages relaxo and plyr with tuning parameters:

• Penalty Parameter (lambda)

• Relaxation Parameter (phi)

Relevance Vector Machines with Linear Kernel (method = 'rvmLinear')

For regression using package kernlab with no tuning parameters

Relevance Vector Machines with Polynomial Kernel (method = 'rvmPoly')

For regression using package kernlab with tuning parameters:

• Scale (scale)

• Polynomial Degree (degree)

Relevance Vector Machines with Radial Basis Function Kernel (method = 'rvmRadial')

For regression using package kernlab with tuning parameters:

• Sigma (sigma)

Ridge Regression (method = 'ridge')

For regression using package elasticnet with tuning parameters:

• Weight Decay (lambda)

Ridge Regression with Variable Selection (method = 'foba')

For regression using package foba with tuning parameters:

• Number of Variables Retained (k)

• L2 Penalty (lambda)

Robust Linear Discriminant Analysis (method = 'Linda')

For classification using package rrcov with no tuning parameters

Robust Linear Model (method = 'rlm')

For regression using package MASS with no tuning parameters

Robust Quadratic Discriminant Analysis (method = 'QdaCov')

For classification using package rrcov with no tuning parameters

Robust Regularized Linear Discriminant Analysis (method = 'rrlda')

For classification using package rrlda with tuning parameters:

• Penalty Parameter (lambda)

• Robustness Parameter (hp)

• Penalty Type (penalty)

Robust SIMCA (method = 'RSimca')

For classification using package rrcovHD with no tuning parameters

ROC-Based Classifier (method = 'rocc')

For classification using package rocc with tuning parameters:

• Number of Variables Retained (xgenes)

Rule-Based Classifier (method = 'JRip')

For classification using package RWeka with tuning parameters:

• Number of Optimizations (NumOpt)

Rule-Based Classifier (method = 'PART')

For classification using package RWeka with tuning parameters:

• Confidence Threshold (threshold)

• Confidence Threshold (pruned)

Self-Organizing Map (method = 'bdk')

For classification and regression using package kohonen with tuning parameters:

• Row (xdim)

• Columns (ydim)

• X Weight (xweight)

• Topology (topo)

Self-Organizing Maps (method = 'xyf')

For classification and regression using package kohonen with tuning parameters:

• Row (xdim)

• Columns (ydim)

• X Weight (xweight)

• Topology (topo)

Shrinkage Discriminant Analysis (method = 'sda')

For classification using package sda with tuning parameters:

• Diagonalize (diagonal)

• shrinkage (lambda)

SIMCA (method = 'CSimca')

For classification using package rrcovHD with no tuning parameters

Single C5.0 Ruleset (method = 'C5.0Rules')

For classification using package C50 with no tuning parameters

Single C5.0 Tree (method = 'C5.0Tree')

For classification using package C50 with no tuning parameters

Single Rule Classification (method = 'OneR')

For classification using package RWeka with no tuning parameters

Sparse Linear Discriminant Analysis (method = 'sparseLDA')

For classification using package sparseLDA with tuning parameters:

• Number of Predictors (NumVars)

• Lambda (lambda)

Sparse Mixture Discriminant Analysis (method = 'smda')

For classification using package sparseLDA with tuning parameters:

• Number of Predictors (NumVars)

• Lambda (lambda)

• Number of Subclasses (R)

Sparse Partial Least Squares (method = 'spls')

For classification and regression using package spls with tuning parameters:

• Number of Components (K)

• Threshold (eta)

• Kappa (kappa)

Stabilized Linear Discriminant Analysis (method = 'slda')

For classification using package ipred with no tuning parameters

Stacked AutoEncoder Deep Neural Network (method = 'dnn')

For classification and regression using package deepnet with tuning parameters:

• Hidden Layer 1 (layer1)

• Hidden Layer 2 (layer2)

• Hidden Layer 3 (layer3)

• Hidden Dropouts (hidden_dropout)

• Visible Dropout (visible_dropout)

Stepwise Diagonal Linear Discriminant Analysis (method = 'sddaLDA')

For classification using package SDDA with no tuning parameters

Stepwise Diagonal Quadratic Discriminant Analysis (method = 'sddaQDA')

For classification using package SDDA with no tuning parameters

Stochastic Gradient Boosting (method = 'gbm')

For classification and regression using packages gbm and plyr with tuning parameters:

• Number of Boosting Iterations (n.trees)

• Max Tree Depth (interaction.depth)

• Shrinkage (shrinkage)

Supervised Principal Component Analysis (method = 'superpc')

For regression using package superpc with tuning parameters:

• Threshold (threshold)

• Number of Components (n.components)

Support Vector Machines with Class Weights (method = 'svmRadialWeights')

For classification using package kernlab with tuning parameters:

• Sigma (sigma)

• Cost (C)

• Weight (Weight)

Support Vector Machines with Linear Kernel (method = 'svmLinear')

For classification and regression using package kernlab with tuning parameters:

• Cost (C)

Support Vector Machines with Polynomial Kernel (method = 'svmPoly')

For classification and regression using package kernlab with tuning parameters:

• Polynomial Degree (degree)

• Cost (scale)

• Scale (C)

Support Vector Machines with Radial Basis Function Kernel (method = 'svmRadial')

For classification and regression using package kernlab with tuning parameters:

• Sigma (sigma)

• Cost (C)

Support Vector Machines with Radial Basis Function Kernel (method = 'svmRadialCost')

For classification and regression using package kernlab with tuning parameters:

• Cost (C)

The lasso (method = 'lasso')

For regression using package elasticnet with tuning parameters:

• Fraction of Full Solution (fraction)

Tree Models from Genetic Algorithms (method = 'evtree')

For classification and regression using package evtree with tuning parameters:

• Complexity Parameter (alpha)

Tree-Based Ensembles (method = 'nodeHarvest')

For classification and regression using package nodeHarvest with tuning parameters:

• Maximum Interaction Depth (maxinter)

• Prediction Mode (mode)

Variational Bayesian Multinomial Probit Regression (method = 'vbmpRadial')

For classification using package vbmp with tuning parameters:

• Theta Estimated (estimateTheta)

References

“Using your own model in train” (http://caret.r-forge.r-project.org/custom_models.html)

caret documentation built on May 2, 2019, 5:47 p.m.